Statistical analyses

LS Lisa-Sophie Schoeben
NM Nicole Mohr
CB Corinna Bubak
AS Astrid Schmieder
MS Marthe-Lisa Schaarschmidt
ask Ask a question
Favorite

GraphPad Software (GraphPad Software, Inc., La Jolla, CA 92,037 USA) and IBM SPSS Statistics (Version 25 for Windows, IBM Germany, Ehningen, Germany) were used for analyses. If Gaussian distribution was confirmed with the d’Agostino and Pearson omnibus normality test, statistically significant differences for continuous parameters were examined by unpaired t-tests. For all continuous parameters without Gaussian distribution, the Mann–Whitney test was performed. For Categorial variables the Chi Square test was used as indicated. Analyses of covariance were conducted to compare differences in the outcome variables (t1—t0) between the 2 arms (covariate = baseline values of the respective variable). Data were analyzed by using intention-to-treat analysis. Missing values at t1 were estimated by using “last observation carried forward” (LOCF). Significance was assumed at p < 0.05. Effect size > 0.1 suggests a weak effect, > 0.25 a medium effect, and > 0.4 a strong effect.

Do you have any questions about this protocol?

Post your question to gather feedback from the community. We will also invite the authors of this article to respond.

post Post a Question
0 Q&A